Successful treatment has increased the survival and curability of many CML patients. However, disease recurrence remains a major problem. Relapse from residual leukemia is the most likely mechanism. Because of its extremely high sensitivity, PCR has provided us a unique opportunity to assess the effectiveness of leukemia eradication. Extensive studies by many different investigators have shown that a good quality of remission (persistent PCR negativity suggesting extremely low or free of leukemia burden) can be achieved by successful therapy. As expected, these patients tend to remain in continuous remission for a long time. PCR- positive complete responders, however, appear to have variable clinical outcome: some relapse early, some relapse late and the remaining remain in continuous remission for a long time without any treatment. This observation suggests that the biological activity and the fate of the residual Ph'-positive cells are quite different. Because of its proven usefulness in the detection of minimal residual disease and in the identification of the bcr/abl splicing patterns, we propose to continue employing the PCR assay to assess the therapeutic effectiveness of our newly launched clinical trials and to determine if the bcr/abl splicing patterns influence the therapeutic responsiveness. To elucidate the significance of PCR-positivity in complete cytogenetic responders, we propose to employ clonogenic assays and flow cytometry to determine the self-renewal capability, proliferative activity and immunophenotypic characteristics of the residual Ph'-positive cells. Correlating the results of our new biology-directed PCR assays with the patients' clinical outcome, we propose to identify the important determinants that help predict the clinical outcome of the PCR-positive complete cytogenetic responders.